Análise de séries temporais para avaliação da dinâmica ambiental a partir de dados climáticos e da cobertura da terra nos sertões cearenses

Abstract

The monitoring of droughts in the Brazilian semiarid is of paramount importance for planning and management of natural resources by decision makers. The predominant vegetation in the region is adapted to water stress conditions becoming fundamental from the socio-environmental point of view being a source of raw materials, besides being important for the cycles of water and carbon. The use of cloud platforms such as the Google Earth Engine platform has allowed researchers and technicians to process large volumes of remote sensing data without the need for a supercomputer. The techniques in Big Earth Observation Data using time series can bring benefits to understand the dynamics from a physical point of view in a climatic region in the context of the caatinga biome, so sensitive to climate change. The case study in the Sertões Cearenses showed the convergence between the literature on the occurrence of extreme drought events and the remote sensor data analyzed in time series from 2000 to 2020. Mann-Kendall trend analyses show that time series of precipitation and falling vegetation index and temperature variables with increasing trend may indicate effects of climate change in the region. In addition, the maps generated from Random Forest machine learning in dry periods (2004-2005, 2013-2014, 2019-2020 and 2023-2024) showed how regional relief in the Sertões Cearenses can be considered a drought enhancer, affecting the distribution of rainfall and the occurrence of Soil with High degree of Exposure, as well as the fixation of greener vegetation in areas such as the Ibiapaba Plateau and the Massifs present in the region. Therefore, the low-cost research showed the efficiency of complex processing methods in the macro region of Sertões Cearenses. This methodological approach may encourage other researchers to study the vegetation and climate of the region.